I am using HiFiLES to simulate flow over a flat plate. The simulation is stable in subsonic cases. However, as the inlet velocity rises to M>1.05, the residuals always blow up. I tried many ways to overcome this, including reduce the timestep size and using shock capture mehtod, but none of them is working. Here's my mesh and input file.
The plate is 0.5m long and 0.01m thick. and the mesh size near the trp of the plate is 0.0001. I restarted it from a stable M=0.9 case.
I am using HiFiLES to simulate flow over a flat plate. The simulation is stable in subsonic cases. However, as the inlet velocity rises to M>1.05, the residuals always blow up. I tried many ways to overcome this, including reduce the timestep size and using shock capture mehtod, but none of them is working. Here's my mesh and input file.
The plate is 0.5m long and 0.01m thick. and the mesh size near the trp of the plate is 0.0001. I restarted it from a stable M=0.9 case.
Input_plate
equation 0
viscous 1 riemann_solve_type
vis_riemann_solve_type 0 ic_form 1 test_case 0
order 1
dt_type 1 //global
CFL 0.025 dt 0.0000005 n_steps 2000000 adv_type 3
tau 1.0 pen_fact 0.5
Large Eddy
LES 1 filter_type 2 filter_ratio 2. SGS_model 0 wall_model 0 wall_layer_thickness 0.0025 // slip-wall
Shock Capturing
artif_only 0 artif_type 1 ArtifOn 1 epsilon0 0.005 //useless s0 0.5 kappa 0.0 //useless
Boundary conditions
Char 1 P_Char 101325. Mach_Char 0.1
Sup_In 1 P_Sup_In 101325. Mach_Sup_In 1.05
Sub_Out 1 P_Sub_Out 101325.